1. Field of the Invention
This invention relates generally to communication systems, and, more particularly, to wireless communication systems
2. Description of the Related Art
Service providers typically provide numerous voice and/or data services to subscribers using one or more wired and/or wireless communication systems. Exemplary services include cellular telephony, access to the Internet, gaming, broadcasting or multicasting of audio and/or video, teleconferencing, multimedia programming, and the like. Mobile subscriber units such as cell phones, personal data assistants, smart phones, pagers, text messaging devices, global positioning system (GPS) devices, network interface cards, notebook computers, and desktop computers may access the services provided by the communication systems over an air interface with one or more base stations. Communication between mobile units and base stations are governed by various standards and/or protocols, such as the standards and protocols defined by the 3rd Generation Partnership Project (3GPP, 3GPP2).
Users can negotiate subscriptions and/or service plans with the service providers. A typical service plan is separated into different levels or tiers. For example, a user can subscribe to a basic level that allows a user access to a basic level of bandwidth, a certain amount of data, and/or a particular quality-of-service (or best effort service) for the services provided by the wireless communication system. Each user pays a basic flat rate monthly fee for the basic level of service. Users that expect to use more than the basic level of bandwidth or data, or require a higher quality-of-service, may subscribe to higher level plans. For example, a user that plans to spend a significant amount of time playing online games or using videoconferencing services may subscribe to a gold service plan that provides a higher level of bandwidth, data, and/or quality-of-service than the basic level. Users pay a higher premium price to subscribe to the higher level service plans.
The capacity of the wireless communication system can vary significantly over time and at different locations. Service providers therefore have difficulty predicting the actual bandwidth, throughput, and/or quality-of-service available to subscribers to the different tiers. Consequently, users' expectations are not always met by the current multi-tier service plans. For example, a user that pays a premium to subscribe to a gold service plan may expect seamless and uninterrupted service even when using services (such as gaming and video teleconferencing) that require significant resources such as bandwidth. Premium users may therefore be frustrated by interruptions and/or degraded quality when the user attempts to use resource-intensive applications at times or in locations where the required resources are scarce, e.g., due to low capacity of the system and/or high usage by other subscribers. These frustrated users may feel that they are not getting good value and may consider dropping the gold service plan and perhaps even switching service providers.
Service providers also have difficulty predicting resource usage of the different subscribers, at least in part because of the wide variety of services available to each subscriber and the idiosyncratic choices made by each subscriber. For example, one subscriber to a gold service plan may use a mobile phone exclusively for voice communication and may therefore use significantly fewer resources than another gold service plan subscriber that uses a smart phone for online gaming.
The inability of service providers to predict individual resource usage can reduce the actual capacity of the wireless communication system. For example, a conventional admission control algorithm assumes that each user requesting access to the system will utilize a predetermined amount of system resources, which may be referred to as a resource margin. The admission control algorithm will admit the requested call if the system can provide the assumed resource margin and will reject the requested call if it determines that it does not have sufficient resources to support the requested call. However, the estimated margin can be very inaccurate when the actual resource consumption of a particular user differs from the expected average resource usage. Admission control algorithms typically assume a relatively high (or worst-case) resource margin and so they tend to overestimate the resources required to support requested calls. Consequently, system capacity may be erroneously reduced, e.g., because calls that have relatively low resource consumption may be rejected because the resource margin for these calls has been overestimated.